Speech recognition system, speech recognition method, and program

Abstract

A speech recognition system according to one embodiment includes: a speech recognition control part configured to determine whether or not to execute speech recognition on a real-time basis, on voice data that is obtained from a voice telephone call; a speech recognition part configured to execute the speech recognition on the voice data when execution of the speech recognition on a real-time basis is determined, and generate text that represents an outcome of the speech recognition; and a UI providing part configured to display a screen, on which the generated text can be referenced on a real-time basis, on a terminal connected via a communication network, and the speech recognition control part is further configured to determine, when the screen is displayed on the terminal, executing the speech recognition on a real-time basis, on voice data being a source of text that can be referenced on the screen.

1 . A speech recognition system comprising:
 at least one processor; and   memory storing instructions that, when executed by the at least one processor, cause the speech recognition system to perform operations, the operations comprising:
 determining whether to execute speech recognition on a real-time basis, on voice data, wherein the voice data is obtained from a voice telephone call; 
 executing, the speech recognition on the voice data when a result of the determining whether to execute speech recognition indicates execution of the speech recognition on a real-time basis; and 
   generating text that represents an outcome of the speech recognition;
 displaying a screen, wherein the generated text is referenced on a real-time basis, from a terminal connected to the speech recognition system via a communication network; and 
 when the screen is displayed on the terminal, executing the speech recognition on a real-time basis, on voice data, wherein the voice data represents a source of text referenced on the screen. 
   
     
     
         2 . The speech recognition system according to  claim 1 , wherein the operations further comprises:
 determining, at predetermined intervals of time, whether a computing resource is available for use for the speech recognition, and, when the unoccupied resources are available, executing the speech recognition on voice data, wherein the result of the determining whether to execute the speech recognition indicates the speech recognition on a non-real-time basis, and   executing the speech recognition on the voice data, for which the execution of the speech recognition on a real-time basis has not been determined, and generating text that represents an outcome of the speech recognition.   
     
     
         3 . The speech recognition system according to  claim 2 , wherein the set of operations further comprises:
 selecting, randomly or according to predetermined rules, one or more pieces of voice data for executing the speech recognition, from among the voice data, for which the execution of the speech recognition on a real-time basis has not been determined; and   executing the speech recognition on the one or more pieces of voice data selected, and generate text that represents an outcome of the speech recognition.   
     
     
         4 . The speech recognition system according to  claim 2 , wherein the set of operations further comprises:
 when the speech recognition is determined to be executed on voice data on a real-time basis, determining whether or not the unoccupied resources are available;   when the unoccupied resources are not available, selecting one or more pieces of voice data, for which the speech recognition is cancelled, among the voice data, for which the execution of the speech recognition on a real-time basis has not been determined; and   cancelling the speech recognition for the selected one or more pieces of voice data, for which the speech recognition is cancelled.   
     
     
         5 . The speech recognition system according to  claim 4 , wherein the set of operations further comprises selecting, when the unoccupied resources are not available, the one or more pieces of voice data, for which the speech recognition is cancelled, among the voice data, for which the execution of the speech recognition on a real-time basis has not been determined, randomly or according to predetermined rules. 
     
     
         6 . The speech recognition system according to  claim 1 , wherein the set of operations further comprises displaying the screen on one or both of a terminal that a first user engaged in the voice telephone call uses, and a terminal that a second user monitoring the voice telephone call of the first user uses. 
     
     
         7 . The speech recognition system according to  claim 1 , wherein the set of operations further comprises:
 storing telephone call data related to the voice telephone call;   searching for telephone call data stored in the storage part based on search conditions specified by the terminal and;   executing, when the searched telephone call data is displayed on the terminal and the speech recognition has not been completed with respect to voice data corresponding to the searched telephone call data, the speech recognition on the voice data.   
     
     
         8 . The speech recognition system according to  claim 1 , wherein the set of operations further comprises dividing the voice data into predetermined voice activity units and executing the speech recognition, in parallel, per voice activity unit. 
     
     
         9 . A speech recognition method to be executed on a computer, the method comprising:
 determining whether or not to execute speech recognition on a real-time basis, on voice data that is obtained from a voice telephone call;   executing the speech recognition on the voice data when execution of the speech recognition on a real-time basis is determined, and generating text that represents an outcome of the speech recognition;   displaying a screen, on which the generated text can be referenced on a real-time basis, on a terminal connected via a communication network; and   determining, when the screen is displayed on the terminal, executing the speech recognition on a real-time basis, on voice data being a source of text that can be referenced on the screen.   
     
     
         10 . A computer-readable non-transitory recording medium storing a program that, when executed on a computer, causes the computer to perform the speech recognition method of  claim 9 .